油气管道智能化运行解决方案的思考  

Thinking on the solution of intelligent oil and gas pipeline operation

作　　者：徐波[1] 李博[2] 宋小晖 兰浩[1] 段林杰 XU Bo;LI Bo;SONG Xiaohui;LAN Haol;DUAN Linjie(PetroChina Pipeline Company;CNPC Key Laboratory of Oil ＆ Gas Storage and Transportation,PetroChina Pipeline R ＆ D Center;China Petroleum Pipeline Engineering Co.Ltd.)

机构地区：[1]中国石油管道公司 [2]中国石油管道科技研究中心中国石油天然气集团公司油气储运重点实验室 [3]中国石油管道局工程有限公司

出　　处：《油气储运》2018年第7期721-727,共7页Oil & Gas Storage and Transportation

摘　　要：基于管道行业自身业务特点,明确其智能化运行实质,寻找有效方法,提升管道行业的智能化水平,最终实现全网优化运行已经成为油气储运工程领域亟待解决的关键技术难题。通过深入分析其他工业领域的智能化实践,明确其共性,结合管道运输行业自身特点,提出了管道智能化运行的内涵。针对如何将智能化技术深度融合于实际管道运行中的问题,以用户用气规律预测为例,提出了基于深度学习和时间序列分析的方法,提高了预测精度;提出了基于神经网络算法,通过与现场运行数据相结合,为设备故障诊断提供了有效的工具;通过优化算法,实现了PID调节的突破。基于上述,提出建设虚拟的油气管道智能化测试平台,将信息流与管道业务深度融合,并引入智能算法,为管道智能化运行提供解决方案。The key technical difficulty that should be solved quickly in the field of oil and gas storage and transportation engineering is to define the essence of intelligent pipeline operation and search for effective methods based on its own business characteristics to enhance the intelligent level of the pipeline industry and ultimately realize optimal operation of the whole pipeline network. In this paper, the intelligentization practice in other industrial fields was analyzed deeply and their commonalities were clarified. Then, the connotation of intelligent pipeline operation was put forward based on the characteristics of the pipeline transportation industry. Next, the prediction on users＇ gas consumption laws was taken as an example to figure out how to apply deeply the intelligent technology into the actual pipeline operation. A method based on deep learning and time series analysis was proposed and the prediction accuracy was improved. An effective tool for equipment fault diagnosis was provided based on the neural network algorithm, combined with field operation data. And breakthrough was realized in PID regulation by means of optimization algorithm. And accordingly, it was proposed to construct the intelligent virtual test platform for oil and gas pipelines, integrate the information flow with the pipeline business deeply, and introduce the intelligence algorithms to provide a solution of intelligent pipeline operation.

关 键 词：油气管道 智能化运行 数据全景化监控 深度学习 智能化虚拟平台 神经网络 

分 类 号：TE832[石油与天然气工程—油气储运工程]